CN111399910B

CN111399910B - User instruction processing method and device

Info

Publication number: CN111399910B
Application number: CN202010172526.3A
Authority: CN
Inventors: 王世镇; 刘思叶; 周志衡; 李智; 沈中南; 徐鹏; 振天; 董敏亚; 余睿
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2020-03-12
Filing date: 2020-03-12
Publication date: 2022-06-07
Anticipated expiration: 2040-03-12
Also published as: CN111399910A

Abstract

The embodiment of the specification provides a processing method of a user instruction, and an execution main body of the method is a client. The method comprises the following steps: firstly, a first session is created in response to an instruction sent by a user to create the session in a first service; secondly, responding to a voice signal input by a user, and generating a voice instruction number to be distributed; then, acquiring a voice operation instruction analyzed by using the voice signal from a server, and allocating the voice instruction number to the voice operation instruction; then, acquiring a first instruction number corresponding to a first operation instruction executed latest in the first session; then, judging whether the voice instruction number is a prior generation number relative to the first instruction number; and finally, under the condition that the number is generated in advance, discarding the voice operation instruction. In this way, conflict management for user instructions may be achieved.

Description

User instruction processing method and device

Technical Field

The embodiment of the specification relates to the technical field of computer processing, in particular to a method and a device for processing a user instruction.

Background

More and more enterprises and structures launch products supporting voice interaction, such as voice ordering systems and the like. Such products allow users to interact in a voice interactive manner, such as ordering and paying. In addition, some products support interaction in other modes, such as a touch screen mode, with the user while performing voice interaction with the user. For the products, because multiple interaction modes are supported simultaneously, the situation that the products receive user instructions is complex, and the processing of the instructions is undoubtedly challenged.

Therefore, a scheme is urgently needed, and aiming at a scene supporting multiple interaction modes, the received user instruction can be processed more accurately and effectively, so that the user experience is improved.

Disclosure of Invention

One or more embodiments in this specification provide a method and an apparatus for processing a user instruction, where conflict management is performed on a received user instruction, so that an execution result of the user instruction is closer to an actual intention of a user, or a confidence level close to an actual intention of the user is higher, thereby improving user experience.

In a first aspect, a method for processing a user instruction is provided, where an execution subject of the method is a client, and the method includes: creating a first session in response to an instruction issued by a user to create a session in a first service; responding to a voice signal input by a user, and generating a voice instruction number to be distributed; acquiring a voice operation instruction analyzed by using the voice signal from a server, and allocating the voice instruction number to the voice operation instruction; acquiring a first instruction number corresponding to a first operation instruction executed latest in the first session; judging whether the voice instruction number is a prior generation number relative to the first instruction number; and in the case of judging that the number is generated in advance, discarding the voice operation instruction.

In one embodiment, prior to fetching the first instruction number, the method further comprises: in response to receiving the first operation instruction, generating the first instruction number assigned to the first operation instruction, the first operation instruction being input through an interaction mode other than voice interaction; and sending the first operation instruction to a server so that the server executes the first operation instruction.

In a specific embodiment, the other interaction mode is a touch screen interaction mode, and the first operation instruction is a touch screen operation instruction; or, the other interaction mode is a physical key interaction mode, and the first operation instruction is a physical key instruction.

In a specific embodiment, in response to starting to collect a voice signal input by a user, generating a voice instruction number to be assigned includes: acquiring a second instruction number corresponding to the instruction number which is generated latest in the first conversation before the voice signal is collected; generating the voice instruction number which is larger than the second instruction number; wherein, in response to receiving the first operation instruction, generating the first instruction number assigned to the first operation instruction comprises: acquiring a third instruction number corresponding to the instruction number which is generated latest in the first conversation before the first operation instruction is received; generating the first instruction number larger than the third instruction number; wherein determining whether the voice command number is a previously generated number relative to the first command number comprises: judging whether the voice instruction number is smaller than the first instruction number or not; wherein, under the condition that the number is generated in advance, abandoning the voice operation instruction comprises: and under the condition that the voice instruction number is judged to be smaller than the first instruction number, the voice operation instruction is abandoned.

In one embodiment, discarding the voice operation instruction in case it is determined that the number is not generated previously comprises: judging whether the contents of the voice operation instruction and the first operation instruction are the same; and under the condition that the voice operation instruction is judged to be the same, discarding the voice operation instruction.

In one embodiment, the method further comprises: under the condition that the number is not generated in advance, informing the server side to execute the voice operation instruction; and receiving an execution result aiming at the voice operation instruction from the server.

In a specific embodiment, the first service is a shopping service, and the voice operation instruction comprises commodity information and an operation type for a commodity shopping cart; wherein after creating the first session, the method further comprises: sending a notification of creating the first session to the server, so that the server creates a shopping cart for the first session; wherein receiving, from the server, an execution result for the voice operation instruction includes: and receiving an updating result aiming at the shopping cart from the server, wherein the updating result is obtained by executing business logic corresponding to the operation type according to the commodity information and updating the shopping cart.

In a more specific embodiment, receiving, from the server, an execution result for the voice operation instruction, further includes: and receiving commodity recommendation information from the server.

On the other hand, in a more specific embodiment, after notifying the server to execute the voice operation instruction and before receiving an execution result for the voice operation instruction from the server, the method further includes: receiving a plurality of matching commodities from the server, the plurality of matching commodities being obtained by searching a commodity library of the first service by using the commodity information; and sending a first matched commodity selected by the user from the plurality of matched commodities to the server, so that the server updates the shopping cart by executing the business logic according to the first matched commodity.

In one embodiment, before generating the voice instruction number to be assigned in response to starting to collect the voice signal input by the user, the method further includes: in response to the creation of the first session, beginning to detect whether a valid voice signal is present in an environment in which the client is located, the valid voice signal corresponding to the voice signal input by the user; wherein, in response to starting to collect the voice signal input by the user, generating a voice instruction number to be assigned, comprises: under the condition that effective voice signals exist, continuously collecting streaming voice signals input by a user, and generating the voice instruction numbers; the acquiring of the voice operation instruction analyzed by using the voice signal from the server includes: based on the streaming voice signal, continuously calling sentence breaking service provided by the server so that the server converts the streaming voice signal into a corresponding recognition text at each moment and judges whether to break sentences after the recognition text at the moment until a sentence breaking success notice is received; and receiving the voice operation instruction from the server, wherein the voice operation instruction is obtained by analyzing the newly added sentence fragment corresponding to the sentence break success notice.

In a specific embodiment, after the client receives a new sentence break text corresponding to a successful sentence break, the method further includes: and under the condition that the effective voice signal is still detected, continuing to collect the subsequent streaming voice signal input by the user and generating other voice instruction numbers to be distributed.

In a second aspect, an apparatus for processing a user instruction is provided, the apparatus being integrated in a client, the apparatus comprising: a session creating unit configured to create a first session in response to an instruction to create a session in a first service issued by a user; the voice number generating unit is configured to respond to the voice signal input by the user, and generate a voice instruction number to be distributed; the voice instruction acquisition unit is configured to acquire a voice operation instruction analyzed by using the voice signal from a server; a voice number assigning unit configured to assign the voice instruction number to the voice operation instruction; a first number acquiring unit configured to acquire a first instruction number corresponding to a first operation instruction executed most recently in the first session; a number judgment unit configured to judge whether the voice instruction number is a previous generation number with respect to the first instruction number; and the instruction processing unit is configured to abandon the voice operation instruction when judging that the number is generated in advance.

In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed in a computer, causes the computer to perform the method provided in the first aspect.

In a fourth aspect, a computing device is provided, comprising a memory and a processor, wherein the memory has stored therein executable code, and the processor, when executing the executable code, implements the method provided in the first aspect.

In summary, by using the method and apparatus for processing a user instruction disclosed in the embodiments of the present specification, through performing conflict management on a received user instruction, an execution result of the user instruction can be closer to an actual intention of a user, or a confidence degree of the user instruction closer to the actual intention of the user is higher, so as to improve user experience.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments disclosed in the present specification, the drawings needed to be used in the description of the embodiments will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments disclosed in the present specification, and it is obvious for those skilled in the art to obtain other drawings based on the drawings without creative efforts.

FIG. 1 illustrates a usage scenario diagram for a client supporting multiple interaction modes, according to one embodiment;

FIG. 2 illustrates a flow diagram of a method of processing a user instruction according to one embodiment;

FIG. 3 illustrates a schematic diagram of the interaction between a client and a server, according to one embodiment;

FIG. 4 shows a schematic diagram of the interaction between a client and a server, according to another embodiment;

FIG. 5 illustrates a block diagram of a processing device for user instructions, according to one embodiment.

Detailed Description

Embodiments disclosed in the present specification are described below with reference to the accompanying drawings.

As described above, for a product supporting multiple interaction modes (or interaction in multiple modes) at the same time, it is complicated to receive a user instruction. The plurality of interaction modes may include a voice interaction mode, a touch screen interaction mode and a physical key interaction mode. Generally, the user instructions may be executed sequentially in the order of their reception, however, the voice operation instruction (hereinafter, or simply, voice instruction) input by the user through the voice interaction mode has a problem of time delay. Specifically, for a voice instruction, a voice signal input by a user needs to be analyzed to obtain a voice instruction, and then the voice instruction can be executed, and for an operation instruction input through other interaction modes, such as a touch instruction, a physical key instruction, and the like, the operation instruction can be directly executed locally after being received, or can be immediately sent to a remote server to be executed.

Further, there is a time delay in the voice command, which may cause a command conflict, for example, when the user starts to input voice, before the corresponding voice command is analyzed, the user also inputs a touch screen command, and the touch screen command is executed, at this time, if the analyzed voice command is executed, it is highly likely that the voice command does not conform to the real intention of the user. For example, some users may be used to speak while touching the screen, and specifically refer to a usage scenario diagram for a client supporting multiple interaction modes shown in fig. 1, at this time, if the received touch screen instruction and the subsequently parsed voice instruction are executed simultaneously, execution will be repeated (for example, a next single 2 hamburgers, and the user only wants to click 1 hamburgers), which not only consumes computing resources, but also may cause the user to need to perform an additional operation (for example, delete 1 hamburger in a shopping cart). For another example, the user intensively and continuously inputs 4 touch screen instructions and 1 voice instruction, the product executes the first 4 touch screen instructions, the voice instruction is not processed, and at this time, the user inputs a touch screen instruction having the same content as the voice instruction for failing to input the voice instruction because the user does not immediately see the execution result of the voice instruction, or changes the idea and inputs other instructions having different contents from the voice instruction, and at this time, if the voice instruction and the touch screen instruction are sequentially executed, the waste of computing resources and poor user experience will also be caused.

Based on the above observation and analysis, the inventor proposes a processing method for a user instruction, in one embodiment, at a time when a user voice input starts to be received, an instruction number to be assigned is generated for a subsequently analyzed voice instruction, and after the voice instruction is analyzed, the instruction number is assigned to the voice instruction, and further, whether to execute the voice instruction is determined based on a sequence of generating the instruction number and an instruction number of a newly executed instruction. Therefore, conflict management on the received user instruction can be realized, the execution result of the user instruction is closer to the actual intention of the user, or the confidence degree of the close to the actual intention of the user is higher, and the user experience is improved.

The following describes specific implementation steps of the method disclosed in the present specification with reference to specific examples.

Specifically, fig. 2 shows a flowchart of a processing method of a user instruction according to an embodiment, and an execution subject of the method is a client, for example, a client of the above product, and for example, a vertical IoT (Internet of Things) tool shown in fig. 1. As shown in fig. 2, the method may include the steps of:

step S210, responding to an instruction sent by a user to establish a session in a first service, and establishing the first session; step S220, responding to the voice signal input by the user, and generating a voice instruction number to be distributed; step S230, acquiring a voice operation instruction analyzed by using the voice signal from a server, and allocating the voice instruction number to the voice operation instruction; step S240, acquiring a first instruction number corresponding to a first operation instruction executed latest in the first session; step S250, judging whether the voice instruction number is a prior generation number relative to the first instruction number; in step S260, when it is determined that the number is generated in advance, the voice operation command is discarded.

With respect to the above steps, it should be firstly explained that the first service mentioned in the above steps and the "first" in the first session, etc., and the subsequent similar terms mentioned in the above steps are only used for distinguishing the same kind of things, and have no other limiting effect.

The steps are as follows:

first, in step S210, a first session is created in response to an instruction issued by a user to create a session in a first service.

Specifically, the client is used for providing a first service to a user. In one embodiment, the first service may be a shopping service or an ordering service. In one particular embodiment, the target items involved in the shopping service may be a variety of items, such as meal items, clothing, pants, shoes, hats, accessories, and the like. In another embodiment, the first service may be an arithmetic service. In a particular embodiment, the arithmetic service may be a service that calculates the total number of legs that a plurality or plurality of animals have. In yet another embodiment, the first service may be a chat service. In one particular embodiment, the chat service may involve a variety of target topics, such as weather, ball games, news, and the like.

In one embodiment, the client may create or start the first session in response to an instruction from any user to begin using the first service, or to create a session in the first service. In a specific embodiment, the instruction to start using the first service may be a click instruction or a voice control instruction. In one example, the click command may correspond to a click operation to start using a service icon (e.g., a start-order icon) in the client interface. In one example, where the voice control command may say "small x, i.e., a main meal" for the user, where "small x" is a voice wake up word.

In the above, the creation of the first session may be realized.

Next, in step S220, in response to the start of the collection of the voice signal input by the user, a voice instruction number to be assigned is generated.

For the above starting to collect the voice signal input by the user, in an embodiment, the method may include: and in response to receiving a voice input instruction sent by a user, starting to collect a voice signal input by the user. In a specific embodiment, the voice input command may be a touch screen click command or a physical key command. In one example, the click command may correspond to a click operation of a voice input icon in the client interface. In one example, the physical key instruction may correspond to a pressing operation of a physical key on the client for activating the voice input function.

In another embodiment, before step S220, the method may further include: in response to the creation of the first session, beginning to detect whether a valid voice signal is present in an environment in which the client is located, the valid voice signal corresponding to the voice signal input by the user. Accordingly, the starting to collect the voice signal input by the user may include: and in the case that the existence of the valid voice signal is detected, continuously collecting the voice signal input by the user. In a specific embodiment, the client may start to continuously collect Voice signals in its environment in response to the creation of the first session, and then perform VAD (Voice Activity Detection) Detection on the collected Voice signals at each time, and when the Detection result is that the energy of the Voice signal at the time is greater than a preset threshold value, determine that a valid Voice signal exists in the Voice signal, or determine that a user Voice input is detected, and accordingly start to collect the Voice signal input by the user.

Therefore, the voice signals input by the user can be collected. Further, in response to the start of the collection, a voice instruction number to be assigned is generated. In an embodiment, a second instruction number may be obtained first, where the second instruction number corresponds to a second operation instruction that is newly generated (or last generated, or a difference between a generation time and a current time is minimum) in the first session before the collection of the voice signal is started; then, in one particular embodiment, a voice instruction number (e.g., 4) greater than the second instruction number (e.g., 3) may be generated, and in another particular embodiment, a voice instruction number (e.g., 67) less than the second instruction number (e.g., 68) may be generated. In another embodiment, the starting time of acquisition for the voice signal may be determined, and then the number corresponding to the acquisition time may be used as the voice instruction number. In one example, assume that the start acquisition time is 13(h):01(min):56(s):054(ms), and thus the voice instruction number can be determined to be 130156054.

The above may be implemented such that the voice instruction number to be assigned is generated in response to the start of the collection of the voice signal input by the user. In one embodiment, the number of the voice command may be one or more, and is used for allocating to one or more voice commands which are possibly analyzed subsequently.

Then, in step S230, a voice operation command analyzed by using the voice signal is obtained from the server, and the voice command number is assigned to the voice operation command.

In one embodiment, the client may send the complete voice signal input by the user to the server, so that the server parses the voice operation instruction from the complete voice signal. In a specific embodiment, after detecting the voice input of the user, the client continues to collect the voice signal until detecting that the voice input of the user stops, and at this time, the voice signal collected during the time and input by the user is sent to the server for the server to analyze. Further, after receiving the voice signal, the server recognizes a Recognition text corresponding to the voice signal by using an ASR (Automatic Speech Recognition) technology, and jointly inputs the audio features extracted from the voice signal and the text features extracted from the Recognition text into a pre-trained instruction Recognition model to obtain one or more corresponding voice operation instructions. Thus, the client can receive the analyzed voice operation instructions from the server.

In another embodiment, considering that there is a case where the user intends to input a plurality of voice commands at a glance, it is desirable to quickly parse and process (e.g., execute) each voice operation command one by one in order to provide the user with a better experience. At this time, the client can continuously call sentence-breaking service provided by the server based on the continuously collected streaming voice signals input by the user, so that the server converts the streaming voice signals into corresponding recognition texts at each moment, and judges whether to perform sentence-breaking after the recognition texts at the moment until a sentence-breaking success notification is received; and receiving the voice operation instruction from the server, wherein the voice operation instruction is obtained by analyzing the newly added sentence fragment corresponding to the sentence break success notice. Further, after receiving the sentence-break success notification, the client may continue to collect subsequent streaming voice signals input by the user and generate other voice instruction numbers to be allocated, when the valid voice signals are still detected. Thus, the user can receive the user voice and analyze the instruction without waiting for the user voice input.

In a specific embodiment, the determining, by the server, whether to perform sentence break after the recognized text at the time may include: firstly, comparing a current recognition text at the current moment with an existing sentence-break text to obtain an incremental text, wherein the existing sentence-break text is a sentence-break fragment sequence formed by carrying out sentence-break on the recognition text before the current moment, for example, assuming that the existing sentence-break text is 'i want a hamburger', the current recognition text is 'i want a hamburger and then a cola', and the incremental text obtained at the moment is 'then a cola'; then, matching the incremental text based on a plurality of pre-acquired keywords corresponding to the first service to obtain a matching word set, for example, assuming that the plurality of keywords include hamburger, cola, chicken wings, and the like, the obtained matching word set may be { cola }; then, based on at least the set of matching words, text features are determined, for example, based on the set of matching words { cola }, the text features can be determined as: an effective trade name number of 1; then, at least based on the text characteristics, judging whether to perform sentence break after the text is currently recognized, so as to determine the incremental text as a new sentence break segment.

Further, the server can analyze the new sentence fragment obtained by sentence break to obtain the analyzed voice operation instruction. In a specific embodiment, firstly, performing word segmentation on the newly added sentence fragment (such as a hamburger and then ice cream) to obtain a plurality of words, and forming a word segmentation sequence (such as a hamburger and then ice cream, wherein different words are separated by a blank space); then, in the word segmentation sequence, a plurality of word segments (such as hamburgers and ice creams) related to the commodity description in the first service are respectively converted into markers (such as T _ XX) in a preset format, and a mixed sequence (such as T _01 and T _02) comprising the word segments and the markers is obtained; then, based on a preset semantic template set, performing loop analysis on the current sequence initially being the above mixed sequence, where any one loop analysis includes, starting from the head of the current sequence (e.g., T _01 and then T _02), truncating a sequence portion (e.g., a T _01) matching with a certain semantic template (e.g., ADD _ pattern _1 ═ quantitygoods) in the multiple semantic templates, obtaining an analyzed operation instruction (e.g., ADD, { "hamburg", 1}) according to an operation type (e.g., ADD) corresponding to the certain semantic template and a product descriptor (e.g., goods) in the sequence portion, and updating the current sequence (e.g., updating to the next T _ 02). Therefore, through multiple rounds of loop analysis until the preamble column is empty, the full amount of user voice commands can be comprehensively, accurately and efficiently analyzed from the newly added sentence fragment.

For ease of understanding, the setting of the semantic template set is described below. Specifically, the semantic template set may be preset by a worker, and specifically, the inventor finds that, no matter how complex and changeable the input content of the user is, the input content of the user may be split into text segments with smaller granularity corresponding to different operation instructions, so that the semantic template set including a plurality of semantic templates may be set according to a large amount of historical or possible user inputs for performing loop analysis on the mixed sequence.

In one embodiment, a plurality of operation types supported in the first service may be determined first. In a particular embodiment, the plurality of operation types may include adding, deleting, clearing, updating, replacing, customizing packages, and the like. It should be noted that clearing, updating, replacing, customizing package and other complex instruction parsing cannot be implemented in a conventional human-computer interaction service, and in the embodiment of the present specification, parsing of these complex operation instructions can be implemented by performing loop parsing using a semantic template set.

Further, in addition to determining the above-mentioned multiple operation types, a plurality of semantic descriptors may be defined according to analysis of a large number of user inputs, and then a corresponding at least one semantic template may be set for each operation type based on the plurality of semantic descriptors. It should be noted that, considering that the operation instruction is performed around the product, the defined multiple semantic descriptors at least include a product descriptor, and in a specific embodiment, the product descriptor corresponds to a combination of the above markers, and in an example, assuming that the first marker in the above predetermined format is T _ xx, the product descriptor may be defined as: goods ═ T _ xx {1,99} (":" >) which indicates that goods can match 1-99T _ xx. In another specific embodiment, the product descriptor corresponds to a combination of a plurality of first markers and terminators, and assuming that the first markers and terminators of the predetermined format are G _ xxx and _ EOS _, respectively, the product descriptor may be defined as: goods ═ (":" < goods ") G _ xxx {1,999} (": ">) _ EOS _, which indicates that goods can match 1-999G _ xxx, ending with _ EOS _. According to a specific example, for the operation type "add", the set at least one semantic template may include: add _ pattern _1 ═ goods, where goods represent item descriptors.

In a specific embodiment, the semantic descriptors may further include a quantity descriptor, and the definition of the quantity descriptor may be: the quantity descriptor is used for matching any number words, including Arabic numerals 1 or 3, or Chinese character form numerals, such as one or one, three or three, and the like. In one example, the quantity descriptor may be denoted as quality. In another specific embodiment, the semantic descriptors may further include a unit descriptor, and the unit descriptor may be defined as: the unit descriptors are used to match any unit word, such as individual, serving, cup, bundle, bag, etc. In one example, the unit descriptor may be expressed as a unit.

In a further specific embodiment, the semantic descriptors may further include a plurality of operation type descriptors, and specifically may include one or more operation type descriptors corresponding to each of the operation types. In a more specific embodiment, the operation type descriptor corresponding to the operation type "add" may include an add verb descriptor (e.g., denoted add _ verbs), an add conjunct descriptor (e.g., denoted add _ linker), and further, in one example, the former may be defined as: add _ verb ═ come "|" buy "|" get "|" dot ", where the symbol" | "represents" or ", the latter being defined as: add _ linker is "and" | "plus" | "and then" | "come again". In another more particular embodiment, the operation type descriptor corresponding to the operation type "update" may include an update verb descriptor (e.g., denoted as update _ verbs), a change verb descriptor (e.g., denoted as change _ verbs), and further, in one example, the former may be defined as: update _ verb ═ about ", the latter is defined as: change _ verb is "to change to" | to change ". In yet another more specific embodiment, the operation type descriptor corresponding to the operation type "delete" may include a former delete verb descriptor (e.g., denoted as delete _ verbjfront), a latter delete verb descriptor (e.g., denoted as delete _ verbjback), and further, in one example, the former may be defined as: delete _ verb _ front ═ delete "|" remove "|" delete ", the latter is defined as: delete _ verb _ back ═ do not want "|" delete "|". In this way, multiple semantic descriptors can be defined.

Based on the plurality of semantic descriptors defined above, a corresponding at least one semantic template may be set for each operation type. In a specific embodiment, for the operation type "add", the set semantic templates may include: add _ pattern1 ═ add _ verbqualityunitgoods, where add _ verb denotes the add verb descriptor above; quality denotes the above-mentioned quantity descriptor, quality? Indicating whether qualification is present or not; unit represents the unit descriptor, unit? Indicating the unit is available or not; goods denote the above commodity descriptors. In a specific embodiment, for the operation type "add", the set semantic templates may further include: add _ pattern2 ═ goods add _ linker quality, where goods represent the commodity descriptor; add _ linker represents the above-mentioned add-conjunction descriptor; quality denotes the above quantity descriptor. In a specific embodiment, for the operation type "delete", the set semantic templates may include: delete _ pattern1 ═ delete _ verb _ front quantitygoods, where delete _ verb _ front denotes the previously deleted verb descriptor; quality denotes the above-mentioned quantity descriptor, quality? Indicating whether qualification is present or not; goods denote the above commodity descriptors. Thus, at least one semantic template can be defined for each operation type, and the semantic templates are further formed.

The semantic template set can be preset, and accordingly, the server can acquire the preset semantic template set, specifically, the acquired semantic template set includes a first definition of a plurality of semantic templates, each semantic template includes a predetermined combination arrangement of a plurality of semantic descriptors and corresponds to an operation type; the plurality of semantic descriptors includes commodity descriptors corresponding to combinations of the markers. In one embodiment, the set of semantic templates further comprises a second definition of the plurality of second descriptors. In an embodiment, the semantic descriptors may further include: a number of operation type descriptors, a quantity descriptor, and a unit descriptor.

By the above method, streaming voice input of a user can be received while user voice instructions are analyzed one by one, so that in a subsequent flow, quick response to the user voice input can be realized by sequentially processing the analyzed voice instructions, further user waiting time is reduced, and user experience is greatly improved.

For one or more analyzed voice commands, after the client obtains the voice command from the server, the client may assign the voice command number to be assigned, generated in step S220, to the analyzed voice commands.

In one embodiment, the number of the generated voice command is 1, and the number of the parsed voice commands is plural, in this case, the number of the voice command may be simultaneously allocated to the plurality of parsed voice commands, that is, the plurality of voice command numbers are allocated to the same command number. In another embodiment, the number of the generated voice instruction is multiple (e.g., 3), the number of the analyzed voice instruction is multiple (e.g., 3), and at this time, the multiple voice instruction numbers may be sorted based on a predetermined sorting manner (e.g., from large to small or from small to large), and then different voice instruction numbers are sequentially allocated to the multiple voice instructions according to the sequence in which the multiple voice instructions are analyzed.

In the above way, the generated voice instruction number to be allocated can be allocated to the analyzed voice operation instruction acquired from the server.

Then, in step S240, a first instruction number is obtained, which corresponds to a first operation instruction executed last in the first session (or last execution, or a difference between an execution time and a current time is minimum).

In one embodiment, before step S240, the method may further include: in response to receiving the first operation instruction, generating the first instruction number assigned to the first operation instruction, the first operation instruction being input through an interaction mode other than voice interaction; and sending the first operation instruction to a server so that the server executes the first operation instruction. In a specific embodiment, the other interaction mode is a touch screen interaction mode, and the first operation instruction is a touch screen operation instruction. In another embodiment, the other interaction mode is a physical key interaction mode, and the first operation command is a physical key command. It should be noted that, for the voice command, the voice command is usually parsed based on the description of the user's accent, the parsing process needs to take a certain time, and the content of the voice command is not as good as that of the touch screen command and the key command, so it is clear, for example, the parsed voice command may be { add, pihamburger, 1}, at this time, since the pihamburger is not a canonical expression of the name of the commodity, further search is needed during execution, there is a possibility that the execution result deviates from the actual intention of the user, and the touch screen command and the key command are issued based on a canonical expression of an operation or a commodity in the screen, so a higher priority can be set for the touch screen command and the key command, that is, the touch screen command and the key command are directly executed after being received.

Further, for the above-mentioned generating the first instruction number assigned to the first operation instruction in response to receiving the first operation instruction, in a specific embodiment, the method may include: firstly, acquiring a third instruction number corresponding to an instruction number which is generated newly in the first session before the first operation instruction is received; then, in correspondence with the above-described generation rule of the voice operation instruction, in a more specific embodiment, a first instruction number (e.g., 3) larger than a third instruction number (e.g., 2) may be generated; in another more specific embodiment, a voice instruction number (e.g., 66) may be generated that is less than the third instruction number (e.g., 67). In another more specific embodiment, the reception time for the first operation instruction may be determined, and then the number corresponding to the reception time may be used as the first instruction number. In one example, the first command number is determined to be 130157000 assuming that the reception time is 13(h):01(min):57(s):000 (ms). In this way, the first instruction number can be generated, on the basis of which it can be retrieved in this step.

Based on the voice command number generated in step S220 and the first command number acquired in step S240, it is determined in step S250 whether or not the voice command number is a previously generated number with respect to the first command number.

In one embodiment, the generation of the voice instruction number includes: acquiring the second instruction number, and generating a voice instruction number larger than the second instruction number, wherein the generation of the first instruction number comprises: and acquiring the third instruction number and generating a first instruction number which is larger than the third instruction number. Based on this, the method can comprise the following steps: it is determined whether or not the voice instruction number is smaller than the first instruction number, and further, in the case where it is determined to be smaller, it is determined to be a previously generated number, and in the case where it is determined to be not smaller (including equal to or larger), it is determined not to be a previously generated number.

In another embodiment, the generation of the voice instruction number includes: acquiring the second instruction number, and generating a voice instruction number smaller than the second instruction number, wherein the generation of the first instruction number comprises: and acquiring the third instruction number and generating a first instruction number smaller than the third instruction number. Based on this, the method can comprise the following steps: it is determined whether or not the voice instruction number is larger than the first instruction number, and further, in the case where it is determined to be larger, it is determined to be a previously generated number, and in the case where it is determined to be not larger (including equal to or smaller), it is determined not to be a previously generated number.

In yet another embodiment, the generating of the voice instruction number includes: determining the starting acquisition time for the corresponding voice signal, and then using the number corresponding to the acquisition time as the voice instruction number, wherein the generation of the first instruction number comprises the following steps: and determining the receiving time of the first operation instruction, and taking the number corresponding to the receiving time as the first instruction number. Based on this, the method can comprise the following steps: it is determined whether or not the voice instruction number is smaller than the first instruction number, and further, in the case where it is determined to be smaller, it is determined to be a previously generated number, and in the case where it is determined to be not smaller (including equal to or larger), it is determined not to be a previously generated number.

In the above, it is possible to determine whether or not the voice command number is a previous generation number with respect to the first command number. Further, in step S260, in the case where it is determined that the number is generated first, the voice operation instruction is discarded. In one embodiment, in the case where it is determined that the number is generated previously, the voice operation instruction is directly discarded, which means that the voice operation instruction is not executed. In this way, repeated execution of user instructions may be avoided. In another embodiment, in the case that it is determined that the number is generated first, it may be further determined whether the contents of the voice operation instruction and the first operation instruction are the same, and then, in the case that it is determined that the contents are the same, the voice operation instruction is discarded, and in the case that it is determined that the contents are not the same, the server is notified to execute the voice operation instruction. Therefore, the accuracy of instruction conflict management can be further improved, and the processing result of the instruction is closer to the real intention of the user.

On the other hand, when the number is judged not to be generated in advance, the server is informed to execute the voice operation instruction, and the execution result of the voice operation instruction is received from the server.

In an embodiment, the first service is a shopping service, and the voice operation command includes commodity information and an operation type for a commodity shopping cart. In a specific embodiment, the merchandise information includes at least one of: commodity attribute, commodity standard name and commodity common name. Based on this, after the above step S210, the method may further include: and sending a notice of creating the first session to the server so that the server creates a shopping cart for the first session. Further, the receiving, from the server, the execution result for the voice operation instruction may include: and receiving an updating result aiming at the shopping cart from the server, wherein the updating result is obtained by executing business logic corresponding to the operation type according to the commodity information and updating the shopping cart.

In a specific embodiment, after the notifying the server to execute the voice operation instruction and before receiving an execution result for the voice operation instruction from the server, the method may further include: receiving a plurality of matching commodities from the server, the plurality of matching commodities being obtained by searching a commodity library of the first service by using the commodity information; and sending a first matched commodity selected by the user from the plurality of matched commodities to the server, so that the server updates the shopping cart by executing the business logic according to the first matched commodity.

In a specific embodiment, the receiving, from the server, an execution result for the voice operation instruction may further include: and receiving commodity recommendation information from the server. Thus, after the conflict management is performed, the voice command is allowed to be executed, and the obtained execution result is more likely to be close to the intention of the user.

In summary, by using the processing method for the user instruction disclosed in the embodiment of the present specification, by performing conflict management on the received user instruction, the execution result of the user instruction can be closer to the actual intention of the user, or the confidence of the actual intention of the user can be higher, so as to improve the user experience.

In addition, in the above embodiments, the creation and state maintenance of the shopping cart and the execution of the user input instruction are mainly described by taking the server side to execute these tasks as an example. In fact, the client can execute the user instruction and update the shopping cart state, at this time, after receiving the touch screen instruction or the physical key instruction, the client can directly execute locally at the client, and in addition, under the condition that the analyzed voice instruction number is not the prior instruction number, the voice instruction corresponding to the voice instruction number can be directly executed locally without being sent to the server for execution.

The above description has been made mainly from the viewpoint of the client as the execution agent, regarding the processing method of the user command. In the following, the processing method is further described from the perspective of interaction between the client and the server, in combination with a specific example.

Specifically, fig. 3 illustrates an interaction diagram between a client and a server according to an embodiment, where the first service provided by the client is a shopping service or a meal ordering service, and the server provides a recognition service for streaming voice signals, a sentence break service for recognizing text, and an instruction parsing service for adding a sentence break segment. As shown in fig. 3, wherein the interaction process may include the steps of:

in step S300, the client receives a creation instruction for creating a session in the first service from the user.

In step S301, the client creates a first session based on the creation instruction.

Step S302, the client sends a creation notification for the first session to the server.

In step S303, the server creates a shopping cart for the first session based on the creation notification.

In step S304, the client generates a voice instruction number to be allocated in response to starting to continuously collect the streaming voice signal input by the user. Specifically, first, a second instruction number (e.g., 0) corresponding to the instruction number newly generated in the first session is acquired, and then, 1 is added to the second instruction number to generate a voice instruction number (e.g., 1).

Step S305, the client continuously calls the streaming voice recognition service and sentence break service provided by the server based on the continuously collected streaming voice signals.

Step S306, the service end carries out voice recognition on the streaming voice signals received at each moment to obtain streaming voice recognition texts at each moment, and carries out sentence break judgment on the streaming voice recognition texts at each moment. In one example, the recognized text "i want" is obtained first and sentence-break is performed, and then the recognized text "i want one" is obtained and sentence-break is performed, and so on.

It should be understood that, in steps S305 and S306, the client collects the streaming voice signal, sends the streaming voice signal, and the server recognizes the streaming voice signal and performs sentence break on the streaming voice recognition text, which is a cyclic process, that is, steps S305 and S306 may be repeatedly and sequentially performed for multiple times until the sentence break is successful at the server, and then steps S320 and S321 are performed for the newly added recognition text corresponding to the successful sentence break.

Step S307, the client receives a touch screen operation instruction and allocates a touch screen instruction number to the touch screen operation instruction. Specifically, first, a third instruction number (e.g., 1) is obtained, which corresponds to the instruction number (e.g., the voice instruction number) generated most recently in the first session, and then, 1 is added to the third instruction number to generate a touch screen instruction number (e.g., 2).

And step S308, the client sends a touch screen operation instruction to the server. For example, the sent touch screen operation instruction is: { add, spicy drumstick, 1 }.

Step S309, the server executes a touch screen operation instruction to update the shopping cart state. For example, add 1 spicy drumstick fort in the shopping cart.

Step S310, the server sends a sentence-break success notice to the client. Further, the client side continues to collect the streaming voice signal and calls a streaming voice recognition service and a sentence breaking service under the condition that the effective voice signal is detected to still exist based on the sentence breaking success notification; and when detecting that no effective voice signal exists, continuously detecting whether a valid voice signal exists.

And step S311, the server analyzes the instruction of the newly added sentence break text to obtain a voice operation instruction. In one example, the parsed voice operation instruction is: { add, spicy drumstick, 1 }.

In step S312, the client receives a voice operation instruction from the server.

In step S313, the client obtains a first instruction number (e.g., 2) corresponding to a first operation instruction (e.g., the touch screen operation instruction) executed most recently in the first session, and determines whether the voice instruction number is a previously generated number relative to the first instruction number.

In step S314, the client discards the voice operation command if it is determined that the number is generated first. For example, the voice instruction number and the first instruction number are 1 and 2, respectively, so that it can be determined that the voice instruction number is a previously generated number, and the voice operation instruction is discarded accordingly.

Specifically, fig. 4 shows an interaction diagram between a client and a server according to another embodiment, where the first service provided by the client is a shopping service or a meal ordering service, and the server provides a recognition service for streaming voice signals, a sentence breaking service for recognizing texts, and an instruction parsing service for adding sentence breaking segments. As shown in fig. 4, wherein the interaction process may include the steps of:

in step S400, the client receives a creation instruction for creating a session in the first service from the user.

In step S401, the client creates a first session based on the creation instruction.

Step S402, the client sends a creation notice aiming at the first session to the server, and sends a commodity list which is currently used for showing to the user in the client.

In step S403, the server creates a shopping cart for the first session based on the creation notification.

Step S404, the client receives a touch screen operation instruction and allocates a touch screen instruction number to the touch screen operation instruction. Specifically, a third instruction number (e.g., 0) is obtained first, which corresponds to the instruction number generated most recently in the first session, and then 1 is added on the basis of the third instruction number to generate a touch screen instruction number (e.g., 1).

Step S405, the client sends a touch screen operation instruction to the server. For example, the sent touch screen operation instruction is: { add, spicy drumstick, 1 }.

In step S406, the server executes a touch screen operation instruction to update the shopping cart state. For example, add 1 spicy drumstick fort in the shopping cart.

In step S407, the client generates a voice instruction number to be allocated in response to starting to continuously collect the streaming voice signal input by the user. Specifically, first, a second instruction number (e.g., 1) is obtained, which corresponds to the instruction number that is newly generated before the streaming voice signal starts to be acquired in the first session (e.g., the instruction number of the touch screen operation instruction described above), and then, 1 is added to the second instruction number to generate a voice instruction number (e.g., 2).

Step S408, the client continuously calls the streaming voice recognition service and sentence break service provided by the server based on the continuously collected streaming voice signals.

Step S409, the service end carries out voice recognition on the streaming voice signals received at each moment to obtain streaming voice recognition texts at each moment, and carries out sentence break judgment on the streaming voice recognition texts at each moment. In one example, the recognized text "still" is obtained first and a sentence break is made, followed by the recognized text "still cup" and a sentence break, and so on.

It should be understood that, in steps S408 and S409, the client acquires the streaming voice signal, sends the streaming voice signal, and the server recognizes the streaming voice signal and performs sentence break on the streaming voice recognition text, which is a cyclic process, that is, steps S408 and S409 may be repeatedly and sequentially performed for multiple times until the sentence break is successful at the server, and then steps S410 and S411 are performed for the newly added recognition text corresponding to the successful sentence break.

And step S410, the server side sends a sentence break success notice to the client side. Further, the client side continues to collect the streaming voice signal and calls a streaming voice recognition service and a sentence breaking service under the condition that the effective voice signal is detected to still exist based on the sentence breaking success notification; and when detecting that no effective voice signal exists, continuously detecting whether a valid voice signal exists.

And step S411, the server side analyzes the instruction of the newly added sentence break text. In one example, the parsed voice operation instruction is: { add, cola, 1 }.

In step S412, the client receives a voice operation instruction from the server.

In step S413, the client obtains a first instruction number (e.g., 1), which corresponds to a first operation instruction (e.g., the touch screen operation instruction) executed most recently in the first session, and determines whether the voice instruction number is a previous generation number relative to the first instruction number.

In step S414, the client sends the analyzed voice operation instruction to the server if it is determined that the serial number is not generated in advance. For example, the voice command number and the first command number are 2 and 1, respectively, so that it can be determined that the voice command number is not a previously generated number, and the server is notified to execute the voice operation command accordingly.

Step S415, the service end updates the shopping cart by executing the business logic corresponding to the operation type according to the commodity information.

In one embodiment, this step may include: retrieving matched commodities matched with the commodity information from the commodity library of the first service; and updating the shopping cart by executing the business logic according to the matched commodity. In a specific embodiment, the retrieving, from the commodity library of the first service, a matching commodity matching the commodity information may include: firstly, acquiring a mapping relation between a pre-established commodity descriptor and a description expansion word, wherein the description expansion word is determined based on commodities contained in a commodity library; then, determining a plurality of commodity descriptors corresponding to the commodity information; then, determining a plurality of description expansion words (such as rattan pepper and Sichuan flavor) corresponding to the commodity description words (such as peppery) based on the mapping relation; and retrieving the matched commodity from the commodity library based on the plurality of commodity description words and the plurality of description expansion words. Further, in a more specific embodiment, the step of retrieving a plurality of matching products may further include: sending the plurality of matching items (such as spicy drumstick burger, rattan pepper drumstick burger and Sichuan drumstick burger) to a client; and receiving a first matched commodity (such as Sichuan-style drumstick) selected from the plurality of matched commodities by the user from the client. Accordingly, the updating the shopping cart by executing the business logic according to the matching commodity may include: and updating the shopping cart by executing the business logic according to the first matched commodity.

In step S416, the server receives the update result for the shopping cart to the client.

Step S417, the client displays the update result to the user.

In step S418, the client receives a session end instruction. In one embodiment, the session end instruction may be user issued. In another embodiment, the session end instruction may be generated after the client countdown times out.

In summary, in the interaction processes shown in fig. 3 and fig. 4, by performing conflict management on the received user instruction, the execution result of the user instruction can be closer to the actual intention of the user, or the confidence of the close actual intention of the user is higher, so as to improve the user experience.

Corresponding to the processing method, the embodiment of the specification further provides a processing device. In particular, FIG. 5 illustrates a block diagram of a processing device for user instructions, the device being integrated into a client, according to one embodiment. As shown in fig. 5, the apparatus 500 includes:

a session creating unit 501 configured to create a first session in response to an instruction issued by a user to create a session in a first service; a voice number generation unit 502 configured to generate a voice instruction number to be assigned in response to start of collection of a voice signal input by a user; a voice instruction obtaining unit 503 configured to obtain, from the server, a voice operation instruction parsed by using the voice signal; a voice number assigning unit 504 configured to assign the voice instruction number to the voice operation instruction; a first number acquiring unit 505 configured to acquire a first instruction number corresponding to a first operation instruction executed most recently in the first session; a number judgment unit 506 configured to judge whether the voice instruction number is a previous generation number with respect to the first instruction number; an instruction processing unit 507 configured to discard the voice operation instruction in a case where it is determined that the number is generated in advance.

In one embodiment, the apparatus 500 further comprises: a first number generation unit 501 configured to generate the first instruction number assigned to the first operation instruction in response to receiving the first operation instruction, the first operation instruction being input through an interaction mode other than voice interaction; a first instruction sending unit 502, configured to send the first operation instruction to the server, so that the server executes the first operation instruction.

In a specific embodiment, the voice number generating unit 502 is specifically configured to: acquiring a second instruction number corresponding to the instruction number which is generated latest in the first conversation before the voice signal is collected; generating the voice instruction number which is larger than the second instruction number; the first number generation unit 505 is specifically configured to: acquiring a third instruction number corresponding to an instruction number which is generated latest in the first session before the first operation instruction is received; generating the first instruction number larger than the third instruction number; the number determination unit 506 is specifically configured to: judging whether the voice instruction number is smaller than the first instruction number or not; the instruction processing unit 507 is specifically configured to: and under the condition that the voice instruction number is judged to be smaller than the first instruction number, the voice operation instruction is abandoned.

In a specific embodiment, the instruction processing unit 507 is further configured to: judging whether the contents of the voice operation instruction and the first operation instruction are the same; and under the condition that the voice operation instruction is judged to be the same, discarding the voice operation instruction.

In one embodiment, the apparatus 500 further comprises: a notification execution unit 508 configured to notify the server to execute the voice operation instruction if it is determined that the number is not generated in advance; an execution result receiving unit 509 configured to receive an execution result for the voice operation instruction from the server.

In a specific embodiment, the first service is a shopping service, and the voice operation instruction comprises commodity information and an operation type for a commodity shopping cart; the apparatus 500 further comprises: a creation notification unit 510 configured to send a notification of creating the first session to the server, so that the server creates a shopping cart for the first session; the execution result receiving unit 509 is specifically configured to: and receiving an updating result aiming at the shopping cart from the server, wherein the updating result is obtained by executing business logic corresponding to the operation type according to the commodity information and updating the shopping cart.

In a more specific embodiment, the execution result receiving unit 509 is further configured to: and receiving commodity recommendation information from the server.

On the other hand, in a more specific embodiment, the apparatus 500 further comprises: a matching product receiving unit 511 configured to receive a plurality of matching products, which are obtained by searching a product library of the first service using the product information, from the server; a selected commodity sending unit 512, configured to send a first matched commodity selected by the user from the plurality of matched commodities to the server, so that the server updates the shopping cart by executing the service logic according to the first matched commodity.

In one embodiment, the apparatus 500 further comprises: a valid voice detection unit 513 configured to start detecting whether a valid voice signal exists in an environment in which the client is located in response to the creation of the first session, the valid voice signal corresponding to the voice signal input by the user; the voice number generating unit 502 is specifically configured to: under the condition that effective voice signals exist, continuously collecting streaming voice signals input by a user, and generating the voice instruction numbers; the voice instruction obtaining unit 503 is specifically configured to: based on the streaming voice signal, continuously calling sentence breaking service provided by the server so that the server converts the streaming voice signal into a corresponding recognition text at each moment and judges whether to break sentences after the recognition text at the moment until a sentence breaking success notice is received; and receiving the voice operation instruction from the server, wherein the voice operation instruction is obtained by analyzing the newly added sentence fragment corresponding to the sentence break success notice.

In a specific embodiment, the apparatus 500 further comprises: and the voice signal acquisition unit 514 is configured to continue to acquire subsequent streaming voice signals input by the user and generate other voice instruction numbers to be allocated in the case that the existence of valid voice signals is still detected.

In summary, by using the processing apparatus for a user instruction disclosed in the embodiments of the present specification, through performing conflict management on a received user instruction, an execution result of the user instruction can be closer to an actual intention of a user, or a confidence of the user instruction closer to the actual intention of the user can be higher, so as to improve user experience.

As above, according to an embodiment of a further aspect, there is also provided a computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method described in connection with fig. 2 or fig. 3 or fig. 4.

According to an embodiment of yet another aspect, there is also provided a computing device comprising a memory having stored therein executable code, and a processor which, when executing the executable code, implements the method described in connection with fig. 2 or fig. 3 or fig. 4.

Those skilled in the art will recognize that, in one or more of the examples described above, the functions described in the embodiments disclosed herein may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium.

The above-mentioned embodiments, objects, technical solutions and advantages of the embodiments disclosed in the present specification are further described in detail, it should be understood that the above-mentioned embodiments are only specific embodiments of the embodiments disclosed in the present specification, and are not intended to limit the scope of the embodiments disclosed in the present specification, and any modifications, equivalent substitutions, improvements and the like made on the basis of the technical solutions of the embodiments disclosed in the present specification should be included in the scope of the embodiments disclosed in the present specification.

Claims

1. A processing method of user instructions, wherein an execution subject of the method is a client, and the method comprises the following steps:

creating a first session in response to an instruction issued by a user to create a session in a first service;

responding to a voice signal input by a user, and generating a voice instruction number to be distributed;

acquiring a voice operation instruction analyzed by using the voice signal from a server, and allocating the voice instruction number to the voice operation instruction;

acquiring a first instruction number corresponding to a first operation instruction executed latest in the first session;

judging whether the voice instruction number is a prior generation number relative to the first instruction number;

under the condition that the serial number is generated in advance, the voice operation instruction is abandoned;

before the first instruction number is acquired, the method further comprises: in response to receiving the first operation instruction, generating the first instruction number assigned to the first operation instruction, the first operation instruction being input through an interaction mode other than voice interaction; and sending the first operation instruction to a server so that the server executes the first operation instruction.

2. The method according to claim 1, wherein the other interaction mode is a touch screen interaction mode, and the first operation instruction is a touch screen operation instruction; or,

the other interaction modes are physical key interaction modes, and the first operation instruction is a physical key instruction.

3. The method of claim 1, wherein generating a voice instruction number to be assigned in response to initiating collection of a voice signal input by a user comprises:

acquiring a second instruction number corresponding to the instruction number which is generated latest in the first conversation before the voice signal is collected;

generating the voice instruction number which is larger than the second instruction number;

wherein, in response to receiving the first operation instruction, generating the first instruction number assigned to the first operation instruction comprises:

acquiring a third instruction number corresponding to an instruction number which is generated latest in the first session before the first operation instruction is received;

generating the first instruction number larger than the third instruction number;

wherein determining whether the voice command number is a previously generated number relative to the first command number comprises:

judging whether the voice instruction number is smaller than the first instruction number or not;

wherein, under the condition that the number is generated in advance, abandoning the voice operation instruction comprises:

and under the condition that the voice instruction number is judged to be smaller than the first instruction number, the voice operation instruction is abandoned.

4. The method of claim 1, wherein discarding the voice operation instruction in the event that it is determined that no number was previously generated comprises:

judging whether the contents of the voice operation instruction and the first operation instruction are the same;

and under the condition that the voice operation instruction is judged to be the same, discarding the voice operation instruction.

5. The method of claim 1, wherein the method further comprises:

under the condition that the number is not generated in advance, informing the server side to execute the voice operation instruction;

and receiving an execution result aiming at the voice operation instruction from the server.

6. The method of claim 5, wherein the first service is a shopping service, and the voice operation instruction comprises commodity information and an operation type for a commodity shopping cart; wherein after creating the first session, the method further comprises:

sending a notification of creating the first session to the server, so that the server creates a shopping cart for the first session;

wherein receiving, from the server, an execution result for the voice operation instruction includes:

and receiving an updating result aiming at the shopping cart from the server, wherein the updating result is obtained by executing business logic corresponding to the operation type according to the commodity information and updating the shopping cart.

7. The method of claim 6, wherein receiving, from the server, a result of the execution of the voice operation instruction, further comprises:

and receiving commodity recommendation information from the server.

8. The method of claim 6, wherein after notifying the server of executing the voice operation instruction and before receiving a result of the execution of the voice operation instruction from the server, the method further comprises:

receiving a plurality of matching commodities from the server, the plurality of matching commodities being obtained by searching a commodity library of the first service by using the commodity information;

and sending a first matched commodity selected by the user from the plurality of matched commodities to the server, so that the server updates the shopping cart by executing the business logic according to the first matched commodity.

9. The method of claim 1, wherein prior to generating a voice instruction number to be assigned in response to initiating collection of a voice signal input by a user, the method further comprises:

in response to the creation of the first session, beginning to detect whether a valid voice signal is present in an environment in which the client is located, the valid voice signal corresponding to the voice signal input by the user;

wherein, in response to starting to collect the voice signal input by the user, generating a voice instruction number to be assigned, comprises:

under the condition that effective voice signals exist, continuously collecting streaming voice signals input by a user, and generating the voice instruction numbers;

the acquiring of the voice operation instruction analyzed by using the voice signal from the server includes:

based on the streaming voice signal, continuously calling sentence breaking service provided by the server so that the server converts the streaming voice signal into a corresponding recognition text at each moment and judges whether to break sentences after the recognition text at the moment until a sentence breaking success notice is received;

and receiving the voice operation instruction from the server, wherein the voice operation instruction is obtained by analyzing the newly added sentence fragment corresponding to the sentence break success notice.

10. The method of claim 9, wherein after the client receives a new sentence break text corresponding to the successful sentence break, the method further comprises:

and under the condition that the effective voice signal is still detected, continuing to collect the subsequent streaming voice signal input by the user and generating other voice instruction numbers to be distributed.

11. A device for processing user instructions, the device being integrated in a client, the device comprising:

a session creating unit configured to create a first session in response to an instruction to create a session in a first service issued by a user;

the voice number generating unit is configured to respond to the voice signals input by the user and generate voice instruction numbers to be distributed;

the voice instruction acquisition unit is configured to acquire a voice operation instruction analyzed by using the voice signal from a server;

a voice number assigning unit configured to assign the voice instruction number to the voice operation instruction;

a first number acquiring unit configured to acquire a first instruction number corresponding to a first operation instruction executed most recently in the first session;

a number judgment unit configured to judge whether the voice instruction number is a previous generation number with respect to the first instruction number;

an instruction processing unit configured to discard the voice operation instruction in a case where it is determined that the number is generated in advance;

the device further comprises:

a first number generation unit configured to generate the first instruction number assigned to the first operation instruction in response to receiving the first operation instruction, the first operation instruction being input through an interaction mode other than voice interaction;

the first instruction sending unit is configured to send the first operation instruction to a server, so that the server executes the first operation instruction.

12. The apparatus according to claim 11, wherein the other interaction mode is a touch screen interaction mode, and the first operation instruction is a touch screen operation instruction; or,

13. The apparatus according to claim 11, wherein the speech number generation unit is specifically configured to:

wherein the first number generation unit is specifically configured to:

wherein the number judging unit is specifically configured to:

wherein the instruction processing unit is specifically configured to:

14. The apparatus of claim 11, wherein the instruction processing unit is further configured to:

15. The apparatus of claim 11, wherein the apparatus further comprises:

the notification execution unit is configured to notify the server to execute the voice operation instruction under the condition that the number is not generated in advance;

an execution result receiving unit configured to receive an execution result for the voice operation instruction from the server.

16. The apparatus of claim 15, wherein the first service is a shopping service, and the voice operation instruction comprises commodity information and an operation type for a commodity shopping cart; the device further comprises:

a creating notification unit configured to send a notification of creating the first session to the server, so that the server creates a shopping cart for the first session;

the execution result receiving unit is specifically configured to:

17. The apparatus of claim 16, wherein the execution result receiving unit is further configured to:

and receiving commodity recommendation information from the server.

18. The apparatus of claim 16, wherein the apparatus further comprises:

a matching product receiving unit configured to receive a plurality of matching products from the server, the plurality of matching products being obtained by searching a product library of the first service using the product information;

and the selected commodity sending unit is configured to send a first matched commodity selected from the plurality of matched commodities by the user to the server, so that the server updates the shopping cart by executing the business logic according to the first matched commodity.

19. The apparatus of claim 11, wherein the apparatus further comprises:

a valid voice detection unit configured to start detecting whether a valid voice signal exists in an environment where the client is located in response to the creation of the first session, the valid voice signal corresponding to the voice signal input by the user;

the voice number generation unit is specifically configured to:

the voice instruction acquisition unit is specifically configured to:

20. The apparatus of claim 19, wherein the apparatus further comprises:

and the voice signal acquisition unit is configured to continue to acquire the subsequent streaming voice signal input by the user and generate other voice instruction numbers to be distributed under the condition that the effective voice signal is still detected.

21. A computer-readable storage medium, on which a computer program is stored, wherein the computer program, when executed in a computer, causes the computer to perform the method of any of claims 1-10.

22. A computing device comprising a memory and a processor, wherein the memory has stored therein executable code that when executed by the processor implements the method of any of claims 1-10.